BALTIMORE – A team of researchers from the University of Chicago has developed a new method that can detect and quantify N6-methyladenosine (m6A) in the transcriptome using enzyme-assisted adenosine deamination.
Described in a study published in Nature Biotechnology earlier this month, the method, named TadA-assisted N6-methyladenosine sequencing (eTAM-seq), offers scientists a new way to investigate m6A modifications in mRNA at a high resolution while lowering the sample input requirement.
"N6-methyladenosine is the most prevalent messenger RNA modification in our body," said Weixin Tang, a chemistry professor at the University of Chicago and one of the corresponding authors of the paper. "You can think it as an important regulatory network encoded in the RNA beyond the genetic sequence."
Despite its functional importance, high-resolution and quantitative detection methods to investigate m6A are still largely lacking, the study's authors noted. Currently, the most widely used strategy to study m6A is through antibody-mediated immunoprecipitation, where antibodies that can recognize m6A are deployed to enrich the mRNA fragments carrying the modifications.
Although this approach is very robust, Tang said, it comes with downsides, such as requiring a large sample input and the inability to quantitatively detect the modifications.
To address these challenges, Tang's team developed eTAM-seq using a version of hyperactive transfer RNA adenosine deaminase (TadA8.20) — a gene-editing enzyme from Escherichia coli that, according to Tang, was originally evolved by Beam Therapeutics, a Broad Institute spinoff founded by researcher David Liu.
"It was a serendipitous discovery that one of the deaminases that we were working on was very active in mammalian setup," Tang said. "Immediately, it came to my mind that, if this is so effective, we may be able to have a bisulfite equivalent [to help detect m6A]."
Mechanistically, the underlying principle of eTAM-seq is largely comparable to that of bisulfite sequencing of DNA for methylation detection, where unmethylated cytosines are deaminated to uracils while leaving methylated cytosines unaffected. In the case of eTAM-seq, unmethylated adenosines are enzymatically converted into inosines, which are then paired with cytosines during reverse transcription and read as guanosine during sequencing. Meanwhile, m6A will be unchanged by the treatment and sequenced as adenosines.
Because eTAM-seq does not involve immunoprecipitation, Tang said one advantage of the method is that it does not lead to sample loss, lowering the sample input requirement. Another benefit is that it offers researchers quantitative and location readouts of the modifications. Furthermore, with the enzyme treatment as the only extra step on top of a conventional RNA-seq workflow, Tang said eTAM-seq's protocol is relatively easy to carry out.
In their study, the researchers applied TadA8.20 to synthetic RNA probes and found that the enzyme achieved a global A-to-I conversion rate of 99 percent. The group also benchmarked eTAM-seq in the whole transcriptomes of HeLa cells and mouse embryonic stem cells and found that the method can preserve RNA integrity while helping achieve m6A detection from limited input samples. Additionally, for site-specific, deep-sequencing-free m6A quantification, the study reported that eTAM-seq's detection limit can go as low as 10 cells.
"We demonstrated detection from 10 cells in this paper," Tang said. "I personally believe that single-cell detection is within reach."
"What's promising about [eTAM-seq] is that it seems to be an analogous version of bisulfite sequencing that we do for DNA epigenetics," said Christopher Mason, a physiology and biophysics professor at Weill Cornell Medicine who was not involved in the work. In addition, he said the low sample input requirement is another appealing feature of the technology.
Broadly speaking, Mason said there are currently three different strategies to study m6A: enrichment-based methods, direct RNA modification detection using single-molecule sequencing, and deamination-based approaches.
Previously, Mason and his collaborators developed MeRIP-seq, a now widely used approach to study m6A using antibody enrichment. The advantage for such an immunoprecipitation-based strategy, Mason said, is that the protocol tends to be simpler and the analysis can be relatively straightforward. However, the downsides are that the data can often be "noisier," and generally cannot deliver base-level resolution.
Meanwhile, scientists including Mason have also tried to directly profile m6A modifications using single-molecule native RNA sequencing. Although these methods can, in principle, tackle long RNA molecules and detect any modifications, they also usually have the highest input requirement and the highest error rate, Mason pointed out.
Prior to this study, a group of researchers in China developed a method called GLORI, which taps glyoxal and nitrite-mediated deamination of unmethylated adenosines for m6A quantification at single-base resolution.
Commenting on the deamination-based methods in general, Mason said they are normally advantageous for their low input requirement and base-level of resolution. However, depending on how the RNA is treated, some of the methods can also degrade the RNA molecules, Mason said.
Despite the advantages of eTAM-seq, Mason said one potential drawback of the method may be the batch effects of the enzyme. "Enzymes can be very helpful," he pointed out, "but they have batch effects sometimes." Therefore, Mason said he would like to see more samples and sample types being run with eTAM-seq to further evaluate how robust the method is.
Meanwhile, as new methods emerge for m6A profiling, Mason said he also thinks it is necessary for the field to start systemically evaluating them. "I think what's really needed now, frankly, in the field is a real rigorous quantification of the accuracy and reproducibility of these range of methods," he said.
Moving forward, Tang said her team hopes to use eTAM-seq to help further unpack the diagnostic function of m6A.
Additionally, since eTAM-seq was initially inspired by bisulfite sequencing, Tang said she is also interested in exploring the possibility of adapting its framework to develop an enzymatic alternative method for bisulfite sequencing. "The challenge there is whether your enzyme will fully differentiate C versus methylated C," she noted.
New England Biolabs already offers a product named Enzymatic Methyl-seq kit (EM-seq) to help identify 5-mC and 5-hmC using the Illumina platform in lieu of bisulfite sequencing. However, in addition to enzymatic deamination by APOBEC, a family of cytidine deaminases, the product also requires TET2, an oxidase, and oxidation enhancer to protect 5-mC and 5-hmC from deamination.
"Basically, the deaminase itself cannot fully differentiate C and 5mC — you add an enzyme to make 5mC more alien from C, so the deaminase can differentiate them," Tang explained. "My hope is that we can have an enzyme and that will readily differentiate C and methylated C."
Meanwhile, commenting on the emerging single-molecule native RNA sequencing technology for modification identification, Tang said she believes the technology, although still early stage, "should be the future."
"If you have reliable detection on native RNA molecule, not only can you detect m6A but you can also detect many other modifications at the same time," she said. "These capacities are missing from current methods."